SAP Netweaver Monitoring Availability

It is not possible to uniquely define availability with regard to IT components. Availability can mean the existence of a process at operating system level. Other definitions of availability can include, for example, the provision of a service within a certain time, or the average time for performing a user action.

SAP provides different mechanisms for checking availability: The following is common to the mechanisms:

• The availability information is usually of a technical nature.

• Once set up, the availability check is performed periodically and without user interaction.

• The result of the check is reported in the central monitoring system.

You have the following options for monitoring availability:

• You can use Availability Monitoring with CCMSPING to monitor ABAP and Java systems and their instances. With this option, the CCMSPING agent queries the message server in each case about which instances are reported as active. With ABAP systems, you can also have the availability of instances and logon groups monitored using a direct RFC call to the instance itself.

• You can check availability at application level with the Generic Request and Message Generator GRMG. With this option, the central system periodically calls a GRMG application using a URL. The GRMG application performs component-specific checks and returns the result of the checks to the central system.

• You can monitor the existence of the relevant process at operating system level. A check of this type determines a prerequisite for the availability of the component. The check is performed by the SAP programs SAPOSCOL and the agent SAPCCMSR

Execution

Check the Availability monitor for alerts. To open the monitor, proceed as described in Tasks. Depending on the affected nodes, you can react to the alert as follows:

 

Affected MTE	Meaning
Check by CCMSPING	With this monitor, you can perform availability monitoring for selected ABAP and Java systems and their instances.
<SysID> Availability: <SysID> on <Host>Instances: <SysID>	Check that the system or instance is definitely not available by attempting to log on. If this is the case, check the developer trace (dev_trace), if appropriate, and restart the system or instance (depending on the platform you are using, either with the MMC or using a start script).
J2EE Engines: Heartbeat by GRMG	The subtree for each GRMG scenario consists of two subtree types: · Availability of the scenario (Was it possible to perform the check of the components?) with the prefix Selfmonitoring · Availability of the monitored components
GRMG: J2EE <SysID> on <Host> Selfmonitoring: Scenario ...	If an alert occurred in this subtree, it was not possible to perform the monitoring of the J2EE Engine, availability values are therefore not delivered for the subtree below .
GRMG: J2EE <SysID> on <Host> Test: <Component>	If an alert occurred in this subtree, the monitored component is not available.
J2EE Engines: OS Processes	You can use this subtree to determine (sorted by the monitored Java systems) whether the most important processes (dispatchers, servers, and SDM) are running at operating system level. If required processes are not running, first check the log files using the Standalone Log Viewer, and then restart the Engine.